Partisan Self-Stereotyping

International Journal of Communication 11(2017), 603–625 1932–8036/20170005

Copyright © 2017 (Jiyoung Han and Daniel B. Wackman). Licensed under the Creative Commons

Attribution Non-commercial No Derivatives (by-nc-nd). Available at http://ijoc.org.

Partisan Self-Stereotyping:

Testing the Salience Hypothesis in a Prediction of Political

Polarization

JIYOUNG HAN1

DANIEL B. WACKMAN

University of Minnesota–Twin Cities, USA

The goal of this study is to theorize the relationship between the news and political

polarization through a lens of group dynamics. Consistent with the salience hypothesis of

the category fit and category accessibility interaction, we first articulate when and how

news exposure makes news consumers think of themselves as Democrats or

Republicans instead of unique individuals. Drawing on group polarization literature, we

further hypothesize partisan self-stereotyping—an automatic reaction to partisan identity

salience—as a mechanism behind the polarizing effect of partisan conflict-framed news.

Two experimental studies provide a consistent pattern of support for our hypotheses.

The implications of these findings were discussed in comparison with extant studies

testing similar news effects under a different theoretical framework—namely, motivated

reasoning.

Keywords: self-stereotyping, polarization, identity salience, conflict framing, self￾categorization

In the presence of conflict, people favor members of their own group (in-group) over members of

the opposing group (out-group; Tajfel & Turner, 1979). In-group favoritism of this sort trumps personal

goals to be fair or maximize the common good (Tajfel, Billig, Bundy, & Flament, 1971). In addition,

decisions made as group members are known to be more extreme than decisions made as unique

individuals (Abrams & Hogg, 1990; Mackie, 1986; Mackie & Cooper, 1984; Price, 1989). The more intense

intergroup competition grows, the more prominent such group influences become (Turner, Hogg, Oakes,

Reicher, & Wetherell, 1987).

Jiyoung Han: [email protected]

Daniel B. Wackman: [email protected]

Date submitted: 2016–03–02

1 This article has been intensively revised based on Jiyoung Han’s master’s thesis, submitted to the

University of Minnesota–Twin Cities in 2012. This research was supported by a research grant from the

Center for Study of Political Psychology and the KRISS research fund from the School of Journalism and

Mass Communication at the University of Minnesota–Twin Cities. We thank two anonymous reviewers for

their helpful comments and suggestions.

604 Jiyoung Han and Daniel B. Wackman International Journal of Communication 11(2017)

As one of commonly agreed news values in the practice of American journalism (de Vreese,

2012; Neuman, Just, & Crigler, 1992; Price & Tewksbury, 1997), conflict prevails in news coverage of

political events (Cappella & Jamieson, 1997; Lawrence, 2000; Patterson, 1993). As vicarious experience,

partisan conflict-framed news may encourage news consumers to respond to a disputed issue in the news

as the members of the Democratic or Republican Party and show more extreme positions in the direction

of their party. To date, however, very few studies have linked media’s emphasis on partisan conflict to

political polarization in American politics through a lens of group dynamics.

Outside of the political realm, Price (1989) showed the potency of social identity elicited by news

exposure. In his experiment, college participants reading a news story about conflict of opinion between

humanities and hard sciences majors expressed stronger support or opposition in the direction of their

group. No such polarization was observed among participants reading a news story without group conflict.

Drawing on group polarization literature (e.g., Turner et al., 1987), Price argued that polarization between the

two majors should be attributed to the heightened salience of group identity resulted from news exposure.

Along similar lines, many studies acknowledge the role of partisan identity salience in political

polarization (e.g., Druckman, Peterson, & Slothuus, 2013; Iyengar, Sood, & Lelkes, 2012; Levendusky,

2013). However, the nature of identity salience has been poorly discussed in these studies. In particular, it

has been unclear whether media exposure equally makes partisan identity salient to all news consumers

or does so at varying degrees. Without deeper discussion, current literature typically assumes that the

former would be the case and argues that defensive information processing in favor of their political

values is a core factor toward polarization in the American mass public (e.g., G. L. Cohen, 2003;

Druckman et al., 2013; Levendusky, 2013). This line of reasoning, however, warrants further scrutiny

because theories of social identity, on which the aforementioned studies build, argue for the latter: There

are gradations in group identity salience (e.g., Hogg, 2003; Oakes, 1987, 2002; Oakes, Turner, & Haslam,

1991; Turner, Oakes, Haslam, & McGarty, 1994).

To overcome this limitation, in this study we define partisan identity salience consistent with the

literature of social identity. We then test how variability in partisan identity salience is associated with

attitude polarization along party lines. Given the importance of group identity salience in the development

of social identity theories (Oakes, 1987, 2002), putting a special emphasis on this topic would be a

pertinent way to incorporate group dynamics into the studies of mass media effects.

The Salience Hypothesis: Fit × Accessibility Interaction

Inspired by Bruner’s (1957) idea of categorization process, social identity theorists conceptualize

group identity salience as an interaction between a situation and a perceiver (Hogg, 2003; Hogg & Turner,

1987; Oakes, 1987; Oakes et al., 1991; Turner et al., 1994). For example, when an African American

boards an elevator filled with White Americans, this comparative situation elicits the person’s racial

identity. By contrast, if the elevator were filled with African Americans, no salience of racial identity would

occur in the mind of the person. This example exhibits that group identity salience is emergent and

context specific.